The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, which represents a theoretical surface for indirect prompt injection if malicious instructions were embedded in logs or configurations.
Ingestion points: Ingests data from LAS (Log ASCII Standard) files and JSON configuration files in scripts/formation_evaluation.py.
Boundary markers: No explicit delimiter-based safety prompts are used to isolate data from instructions.
Capability inventory: The skill's capabilities are limited to file system read/write operations for data logs and plots. It does not possess network exfiltration, subprocess execution, or dynamic code evaluation capabilities.
Sanitization: Uses standard scientific parsing libraries (lasio, petropy, and json) which treat inputs as data structures rather than executable code.
[EXTERNAL_DOWNLOADS]: The skill utilizes several standard, open-source scientific Python packages for its calculations and data handling.
Evidence: Dependencies identified in SKILL.md include petropy, lasio, and numpy. Code references also suggest usage of matplotlib and pandas for visualization and data exporting.

petropy